Back to skill

Security audit

抖音视频文案提取

Security checks for vulnerabilities and agentic risk

Overview

The skill performs the advertised video transcription task, but it uploads user-selected media to a remote service and has under-disclosed network and download risks users should review before installing.

Install only if you are comfortable sending selected videos and prompts to the GuaiKei service and its returned storage endpoint. Avoid using private, sensitive, or internal-network URLs, and do not submit confidential local videos unless you independently trust the provider's retention and deletion promises.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (3)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/utils/validator.js:3
Finding

Unrestricted URL Fetching Enables Server-Side Request Forgery

Content
View full analysis
this.__opts.maxRedirects) { const err = new Error("Too many redirects"); this.__setState(this.__states.FAILED); this.emit("error", err); return reject(err); } const redirectedURL = /^https?:\/\//.test(response.headers.location) ? response.headers.location : new URL(response.headers.location, this.url).href; this.__isRedirected = true; this.__initProtocol(redirectedURL); this.emit("redirected", redirectedURL, this.url); return this.__start(); } ``` ### Technical Analysis The URL validator checks only whether the input uses HTTP or HTTPS. It does not reject loopback, private, link-local, reserved, multicast, or cloud metadata addresses. It also does not resolve and inspect hostnames before connecting. The downloader follows up to ten redirects without applying a network-address policy to each new target. A public endpoint can ...[truncated 1749 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/utils/helper.js:21
Finding

Unbounded Remote Downloads Permit Disk and Resource Exhaustion

Content
View full analysis
{ const options = { retry: { maxRetries: constants.CREATE_MAX_ATTEMPTS, delay: constants.RETRY_INTERVAL, }, override: { skip: true, skipSmaller: true }, }; let progressLog = ""; ``` ```js // scripts/utils/download.js:526-547 const _fileName = this.__getFileNameFromHeaders(response.headers); this.__filePath = this.__getFilePath(_fileName); this.__fileName = this.__filePath.split(path.sep).pop(); if (fs.existsSync(this.__filePath)) { const downloadedSize = this.__getFilesizeInBytes(this.__filePath); const totalSize = this.__total ? this.__total : 0; if ( typeof this.__opts.override === "object" && this.__opts.override.skip && (this.__opts.override.skipSmaller || downloadedSize >= totalSize) ) { this.emit("skip", { totalSize: this.__total, fileName: this.__fileName, filePath: this.__filePath, downloadedSize: downloadedSize, }); this.__setState(this.__states.SKIPPED); return resolve(true); } } this.__fileStream = fs.createWriteStream(this.__filePath, {}); ``` ### Technical Analysis The download workflow does not define a maximum accepted `Content-Length`, a streaming byte limit, a minimum free-space requirement, or a media-type restriction. The response is streamed directly into a local file. A malicious server can omit `Content-Length`, provide a false length, or send an extremely large or effectively endless response. The client continues writing until the connection ends or another environmental failure occurs. Retry behavior can further increase bandwidth and execution-time consumption ...[truncated 1350 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
scripts/utils/upload.js:11
Finding

Service-Controlled Presigned URL Can Direct File Uploads to Arbitrary HTTPS Hosts

Content
View full analysis
{ const fileStats = fs.statSync(filename); const totalSize = fileStats.size; let uploadedSize = 0; let settled = false; const fileStream = fs.createReadStream(filename); // 复制预签名 headers,并补充 OSS PUT 必需字段 const uploadHeaders = Object.assign({}, headers); // 关键修复:明确 Content-Length,否则 Node 对流式 body 默认使用 chunked // Transfer-Encoding,阿里云 OSS 预签名 PUT 通常会拒绝而返回 400/403 uploadHeaders["Content-Length"] = String(totalSize); if (!uploadHeaders["Content-Type"]) { uploadHeaders["Content-Type"] = "application/octet-stream"; } const options = { host: url.hostname, path: url.pathname + url.search, method: "PUT", headers: uploadHeaders, }; const req = https.request( ``` ### Technical Analysis The `www.guaikei.com` API returns a presigned upload URL and request headers. The client validates only that the URL uses HTTPS; it does not constrain the hostname, port, object-storage provider, account, or permitt ...[truncated 1573 chars]
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (22)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The static finding reports undeclared local file path access checks despite a description centered on remote Douyin transcription and LLM-based text generation. Undeclared local path handling can expose users to unexpected access to host files or broaden the skill’s effective reach beyond the stated remote-only use case.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The README defines automatic command conversion from user natural-language requests such as “总结一下这个视频的核心观点” and “用刚才分析的视频提取所有金句,” but it does not clearly bound where or when these phrases should activate. Because the trigger rule is framed broadly as any user request containing summary/rewrite/extract/analyze intent, it risks unintended invocation from ordinary conversation about videos.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · README.md (reported line 99)May include surrounding context.

md
4. 同时传入文件路径与任务ID,优先执行 `--id`,忽略 `--file`
5. 无自定义 prompt 时,默认完整转录视频全部文字

| 用户自然语言指令                                         | 自动生成命令                                                                                                |
| -------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| 视频提取 https://example.com/video.mp4 中的文字          | `node scripts/video2text/index.js --file "https://example.com/video.mp4"`                                   |
| 把本地 /path/to/your/video.mp4 改成小红书风格的文案      | `node scripts/video2text/index.js --file "/path/to/your/video.mp4" --prompt "改写成小红书风格的文案"`       |

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The description lists broad activation phrases such as '抖音视频分析' and also says the skill should be used for extracting audio copy, scene-description copy, summaries, rewrites, translation, and other transformations. In a manifest description, this wide phrasing can overlap with many ordinary requests about generic video analysis or content creation, making it unclear when this skill should activate versus more specialized tools.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The section states '出现以下任意意图时优先调用本技能' and includes broad intents like summarizing video content, turning material into platform-specific copy, extracting quotes, translation, and interview/course/live-review tasks. Because these are common requests that may not be specific to Douyin or even to transcription, the activation boundary is ambiguous and could cause unintended invocation.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · SKILL.md (reported line 121)May include surrounding context.

md
当用户用自然语言下达指令时,按以下映射生成命令,保证识别与执行一致:

| 用户自然语言指令                                     | 生成的命令                                                                                                  |
| ---------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- |
| 提取 https://example.com/video.mp4 里的文字          | `node scripts/video2text/index.js --file "https://example.com/video.mp4"`                                   |
| 总结这个视频的核心观点 https://example.com/video.mp4 | `node scripts/video2text/index.js --file "https://example.com/video.mp4" --prompt "总结这个视频的核心观点"` |

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file emits user-facing error strings exclusively in Chinese, including request failures and token guidance, with no indication that the skill is Chinese-only or that users can opt into another language. This is a natural-language policy concern because it imposes a specific language on all users without visible choice or justification in the file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The user-facing strings emitted on invalid token configuration are entirely in Chinese, including warnings and operational guidance. This imposes a specific language on users without opt-in or any indication that the skill is intended only for a Chinese-language context, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The comment states that uploaded videos are only used for video-to-text service and are automatically deleted after transcription. However, the code in this function only performs an HTTPS PUT upload to a presigned URL and contains no deletion logic, retention control, or enforcement of restricted downstream use.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This code presents flag descriptions, examples, and runtime messages exclusively in Chinese. Because the file does not offer user opt-in for language selection or explain that the skill is intentionally region-specific, it creates a natural-language locale policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The README requires users to obtain a token via WeChat and presents all operational guidance in Chinese without indicating any language or locale choice. This can constitute a language/locale policy issue when a skill effectively assumes a single language for access and usage without user opt-in or documented regional limitation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The manifest description is entirely in Chinese and presents the skill's functionality without any indication that language is selectable or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy for natural-language violations, forcing a specific language without user opt-in can be a locale-policy issue.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

This JavaScript utility contains user-facing comments, error messages, help text, and examples entirely in Chinese, including CLI help and runtime errors. That creates a locale-specific behavior without offering a language choice or documenting that the tool is intentionally limited to Chinese users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This JavaScript file contains a Chinese-only doc comment and multiple Chinese user-facing status/error messages such as "下载完成" and "下载失败", with no indication that the skill is region-specific or that users can opt into another language. That can violate the language/locale policy because the skill effectively forces a specific language in its natural-language output.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

This file contains user-facing natural-language text exclusively in Chinese in the function documentation, and the runtime log/error strings throughout the file are also Chinese-only. Under the policy, forcing a specific language without user opt-in can be a locale-policy violation unless the constraint is explicitly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The banner string is hard-coded in Chinese, which indicates the skill presents itself in a specific language without any visible opt-in or alternative locale handling. Under the policy rule, forcing a specific language without user choice is a natural-language policy concern.

Content

No source excerpt is available for this finding.